1 of 30

IPUMS: An Introduction

(ip’-umz)

2025 DLI National Training

Alexandra Cooper, Queen’s University

Michael Steeleworthy, Wilfrid Laurier University

2 of 30

FAQ: What is IPUMS an acronym for?

Technically, nothing!

IPUMS originally stood for Integrated Public Use Microdata Series.

In 2016, all of our data projects took on the IPUMS prefix as part of their project name (e.g., IPUMS NHGIS, IPUMS IHGIS, etc.). Since not all these projects are microdata and some have access conditions that limit their usage, it became inaccurate to describe IPUMS as a "public use microdata series".

Since then, we've just gone by IPUMS, and no longer treat it as an acronym.

Find a data error – get a mug https://usa.ipums.org/usa/mug.shtml

3 of 30

What is IPUMS?

4 of 30

What is IPUMS?

  • Archive of harmonized microdata
  • Samples of census & health data
  • Enables research across time and political boundaries
  • Hosted by Minnesota Population Center

Collection

U.S.

International

Census Data

1850s-

1960s- +Historic

Health

1960s-

1980s-

Time Use

1930s-

1960s-

Education

1930s-

n/a

Geodata

1790s-

1990s-

5 of 30

IPUMS holdings by the numbers

icons: flaticon.com

Data from over 157 countries!

2.6 billion records! (rows)

581 Billion cells!

Data since 1790s!

6 of 30

Data availability – all regions

7 of 30

Data availability - historical

8 of 30

Topic availability

  • Demographics
  • Economics
  • Education
  • Health

9 of 30

IPUMS variables are harmonized whenever possible

  • Whenever possible, variables are harmonized for consistency across datasets
    • E.g., LFSTAT in Canada would be matched other Labour force status variables
    • This can mean recording values within a variable to harmonize them with similar variables from similar countries
  • Harmonization works across time, datasets, and even countries
  • Over 65,000 harmonized variables
    • Variables and values have been matched/grouped when possible

10 of 30

Harmonization example: MARSTAT (Marital Status)

Each country may code the variable differently. IPUMS must harmonize this to make comparison possible.

IPUMS

Country A

Country B

Country C

Country D

1 (Single)

  1. Single
  1. Single
  1. Single
  1. Single

2 (Married / In Union)

  1. Married, Religious
  2. Married, Civil
  3. Common-law/Co-hab

  1. Married, Religious
  2. Married, Civil
  3. Common-law
  4. Polygamous
  1. In-union
  1. Married
  2. Re-married

3 (Separated)

  1. Separated Informally
  2. Separated Legally
  3. Divorced
  1. Separated
  2. Divorced
  1. Separated
  1. Separated

4 (Widowed)

  1. Widowed
  1. Widowed
  1. Widowed
  1. Widowed

11 of 30

Harmonized variables vs. source variables

“What about the original variable?”

  • The variables from the original dataset are called source variables
  • Source variable values, codes, and labels are not consistent across samples
    • By harmonizing variables, IPUMS develops consistency
  • Source variables can be downloaded
    • You can get both harmonized and unharmonized datasets.
  • Source variables still undergo a degree of processing to regularize them
    • *Unknowns are recoded, data converted to numeric values, source var’s recoded to have a unique name in the IPUMS db

12 of 30

Using IPUMS

13 of 30

Who should use IPUMS? And why?

  • Researchers conducting comparative work
    • Across time and nations
  • Historical work
  • Consistency of practice and significant curatorial control through the entire collection
    • Curation processes are similar for each dataset
  • IPUMS v. World Bank
    • Data vs. statistics, your use case may benefit from one portal or the other
    • Note: data sources are similar; come from national statistical agencies

14 of 30

Registration

  • Microdata is free but registration is required
  • Must state purpose of your project
  • General terms of use:
    • Cannot redistribute data
    • Academic research only
    • Cite IPUMS
  • Each project has its own terms
  • Must register to access each project

15 of 30

IPUMS International�(censuses and labor* force data)

*American data = American spelling

16 of 30

IPUMS International

17 of 30

IPUMS International – Data available

Contains:

  • Basic demographic information
    • Religion
    • Occupation, industry, income, work status
    • Education
    • Type of housing
    • Disability status
    • Household characteristics
    • Supplemental information for some countries - fertility data, migration data, mortality Data, and GIS boundary files
  • Additional data:
  • Fertility, mortality, and migration
  • Geography and GIS

18 of 30

IPUMS International – Select Samples

19 of 30

IPUMS International – Select variables topic

20 of 30

IPUMS International – Select specific variables

21 of 30

IPUMS International – Extract request

22 of 30

IPUMS International – Downloads

23 of 30

IPUMS Discovery�(aka advanced search)

24 of 30

IPUMS Discovery

  • Search across all IPUMS project data
  • Filter by:
      • Project data
      • Country
      • Year

25 of 30

Search results

26 of 30

Training and Support

27 of 30

Training

Data training exercises - https://www.ipums.org/support/exercises

  • Self-paced, interactive exercises
  • Excel, SPSS, SAS, Stata, R
  • Some exercises are online/SDA; have to log in to access

Online tutorials - https://www.ipums.org/support/tutorials

  • Videos guide users through the using IPUMS sites, creating extracts, and using our online analysis tools.

Teaching Resources - https://www.ipums.org/support/teaching-resources

28 of 30

Support

Support - https://www.ipums.org/support

IPUMS User Forum - https://forum.ipums.org/

  • Forum to ask questions, share insights, and learn

Virtual Office hours

  • Held quarterly
  • Provides regular opportunities for face-to-face, personalized support from data experts in a casual atmosphere

Newsletter

Use it for Good Blog

  • News and Information from IPUMS
  • https://blog.popdata.org/

29 of 30

IPUMS Working Paper Series and Bibliography

Working Paper Series

  • Provides in-depth descriptions of IPUMS methods, features, and applications of data
  • Papers go throuhg a review process before being posted
  • https://www.ipums.org/impact/working-papers

Bibliography

  • Search working papers and other research that uses IPUMS data
  • https://bibliography.ipums.org/

30 of 30

Questions!

Alexandra Cooper coopera@queensu.ca

Michael Steeleworthy

msteeleworthy@wlu.ca

IPUMS User Support

ipums@umn.edu